2

Correct me if I am wrong, but while business analysts are often an important role in software delivery teams, but it seems not the case in the data science field.

Commonly available data scientist skill sets state also importance of story-telling skills and understanding customer's business, but is that enough?

Open Food Broker
  • 381
  • 1
  • 2
  • 13

2 Answers2

3

I believe the integration of Business Analyst(s) into data science team makes great sense.

For a number of reasons:

  • there will be a lot of interaction between business proper and the data science team. BAs are used to acting as an interface layer between IT and business functions. This is much harder than it seems, as IT and (for example) Sales & Marketing people are by nature extremely different. It takes a lot of patience and experience to get them both on the same page. Hint: when talking to Sales or Accounting do not start with an icebreak about the virtues of Open Source!
  • BAs usually have better domain knowledge than DSs. They can be of great help in initial phases of the project as they tend to have a more natural "feeling" for the business (e.g. which outcomes are possible, and which not, are we forecasting a rare event and some under/over sampling is in order, what sort of results are no news and what kind are so surprising that they should be brought to board attention immediately etc.). Data Science types are good at finding the correct answers once a problem is formulated, but asking the right question is arguably more important. And it is a task a BA can help with.
  • Business Analysis is around for longer, so there is a deep and broad pool of experienced people around. Data Science is young but sexy and looks nice in a resume, so it will be easy to persuade a smart guy/gal with strong understanding of the underlying business and good relationship with the key players in the company to become a project member. It is a win win situation.
Jindra Lacko
  • 181
  • 2
  • 8
2

I'm not a business analyst so I guess you'll have to take what I have to say with a pinch of salt. From my understanding, Business Analysts responsibilities are focused around improving processes within a company, for example how certain technologies could be implemented to improve a workflow, they are expected to understand how these technologies might improve the workflow, or a product etc, and manage these improvement projects.

This seems to differ from Data Science on an abstract level in that it is exploring known unknowns ("can our process be improved? what technologies/methods exist that could improve it?"), whereas Data Science is great for exploring unknown unknowns. For example, why is this better for our workflow/product specifically? Data Science is great at throwing up results you don't expect, which is one reason why it is so valuable. I may be wrong in saying this, but Business Analysis seems to be relatively free-form depending on the company and the needs, whereas Data Science has a less subjective methodology.

With this in mind, perhaps Data Scientists could be used to better inform your Business Analysts decisions? But the other way around, BAs could perhaps be used to better inform Data Scientists of business processes, or maybe your BA could focus on improving processes to make life easier for your Data Scientists such as data pipelines in a non-automated environment (example: how can we gather more useful data from our vehicle showrooms? What technologies do the DSs need and how can we implement them?).

Dan Carter
  • 1,732
  • 1
  • 11
  • 26